Modern wireless telecommunications systems are evolving to provide high speed packet data services for users of mobile equipment. One example is an ability to provide internet access to a user of mobile equipment. A wireless system that is rapidly evolving in this direction is a Time Division, Multiple Access (TDMA) system known as the Global System for Mobile Communication (GSM), in particular enhanced versions of GSM known as GSM+, GPRS (General Packet Radio Services) and EGPRS (Enhanced General Packet Radio Services).
The GPRS Release '97 was the first standard to provide (limited) packet data services. However, this standard did not provide a capability for the user to control the bit rate(s) and delays for a packet data connection. In the developing Universal Mobile Telecommunication System (UMTS) packet domain permits several packet data connections to be simultaneously maintained, with different qualities of service. Although there have, at present, been two subsequent GPRS releases since the Release '97, the quality of service concept has remained the same.
The GSM/EDGE radio access network (GERAN) release 5 (or simply R5) provides a new radio access network to the UMTS core network, and is to adopt the same quality of service attributes as used in the existing UMTS.
In the UMTS a data connection between a mobile station (MS), such as a cellular telephone, and the third generation (3G) Serving GPRS Support Node (SGSN), or 3G-SGSN, is identified using the Network Service Access Point Identifier (NSAPI) with which the requested quality of service (QoS) parameters are associated. The data connection is realized by a radio access bearer established by the 3G-SGSN to the radio access network. The radio access bearer identity is the same as the NSAPI. That is, in UMTS the data connection is identified using the NSAPI, which also identifies a radio access bearer. In the radio interface the radio access bearer is realized by one or several radio bearers, each having their own identities. During the radio bearer set-up phase the NSAPI is associated with radio bearers and the radio bearers are associated with a channel. As such, in the UMTS radio access network the channel number/identifier unambiguously identifies the data connection and its quality of service parameters and, hence, there is no need to carry either the NSAPI or radio bearer identity in protocol headers.
However, in GERAN R5 there is no provision to associate a data connection to a (physical) channel. As such, one problem that arises is how to identify a data connection in the radio interface.
A second issue relates to improving the flexibility of the GPRS Radio Link Control/Media Access Control (RLC/MAC) layer. An important distinction between the basic GPRS and the UMTS Radio Access Network (URAN) is that the GPRS MAC multiplexes Logical Link (LL) Protocol Data Units (PDUs), while UMTS multiplexes transport (Radio Link Control or RLC) blocks. In general, GPRS multiplexing is inflexible, and is not suitable for connections having different quality of service requirements.
In EGPRS the same access types as in GPRS are supported to establish the Temporary Block Flow (TBF) in the uplink direction (i.e., from the mobile equipment or station to the network). To accomplish this, a control message used by a GPRS mobile equipment to request a packet channel (Packet Channel Request, 11 bits) is re-used for EGPRS.
With Release 5 the standard RLC and MAC sublayers are required to support multiple data flows with different QoS requirements simultaneously. However, the modifications made to the Release 4 (R4) standard to derive the R5 standard must be backward compatible with R4 (and earlier releases back to R97). That is, different mobile stations (MSs) from different releases (R97-R5) must be able to be multiplexed onto the same Packet Data Channels (PDCHs). Put another way, there is no segregation of traffic between R5 MSs and pre-R5 MSs. This implies that the multiplexing fields Temporary Flow Identity (TFI) and Uplink State Flag (USF) in the headers of the RLC/MAC protocol data units must remain unchanged.
Currently in R97-R4 only one data flow at a time is allowed in the MS, in either the uplink or the downlink direction. This data flow is transported on a Temporary Block Flow (TBF) which is identified by a TFI. That is, the TFI uniquely identifies a MS/TBF pair. In the downlink direction (to the MS) the TFI is used to address a block to the MS, and in the uplink direction (from the MS) the TFI is used to identify the owner (MS) of the incoming data block. On a given PDCH a maximum of 32 TBFs (TFI is 5 bits) are allowed in either the uplink or the downlink directions. The USF is used for the dynamic allocation of uplink resources, while on a downlink PDCH Radio Link Control (RLC) data blocks are appended with a USF (by the Media Access Control (MAC)), the value of which allows a unique MS to send one (or four) data blocks in the uplink direction in predefined blocks on the corresponding uplink PDCH. That is, no other MS is allowed to use these blocks. The USF is 3 bits, thereby enabling dynamic allocation for eight different MSs. This implies only eight different dynamic TBFs.
Introducing multiple flows per MS in the uplink through multiple TBFs would thus imply, in this context, that several considerations be made. First, a MS may have several USFs reserved for it: one for each TBF. As such, dynamic allocation is highly restrained. Second, this further implies that the number of MSs on a given PDCH is also highly restrained (maximum of 32 TBFs on the PDCH). Actually, due to the USF constraint only eight MSs in dynamic allocation can be supported per uplink PDCH. Third, introducing multiple flows per MS further implies that the network (NW) schedules all uplink flows, i.e., the MS cannot perform its own scheduling in the uplink direction, which implies further delays in transmitting data. That is, a MS must obtain authorization from the NW before it is enabled to send a particular TBF, which may be unacceptable depending on the desired QoS of the data flow.
The inventors have realized that all uplink resources are under NW control (fixed allocation, dynamic allocation, extended dynamic allocation), which implies that the NW has knowledge of from which MS it will be receiving a data block at any given point in time. Note should be made of the fact that, because of this NW knowledge, the existing TFI in the uplink data blocks is actually not needed, that is, a given MS is not allowed to transmit when it wishes, but only when it is given permission to transmit.